Comparison of HDR Versions (PS3, 360, DC, CPS2)


We did ask for another patch and expressed why it was needed on Unity.

All Sven said was, “there are no more patches planned at this time”

That was like 2 years ago.

True, I remember something like that back then now that you mention it.

I guess either way Capcom is going to focus on the game that brings in the bigger bag of $$$.

So either way it didn’t matter what our smaller community did.

Even if they did patch the game, realistically, that probably wouldn’t be enough to help revive the game even if it soldered the HDR and ST communities together.

A little thread necro… my method will give lag times that are, on average, about 1/2 a frame lower than the camera/LED method that, for example, NKI used.

CPS-2 Green Turbo 1,2 and 3: 2-3 frames. Average 2.5 frames. (3 frames NKI equivalent)

Xbox360 Turbo 0, PS360 PCB: 4.2-5.4 frames. Average 4.8 frames (5.3 frames NKI equivalent)
Xbox360 Turbo 2 PS360 PCB: 3.5-5.8 frames. Average 4.7 frames (5.2 frames NKI equivalent)
Xbox360 Turbo 4 PS360 PCB: 3.3-5.4 frames. Average 4.5 frames (5 frames NKI equivalent)

PS3 Turbo 0, PS360 PCB: 5.2-6.3 frames. Average 5.7 frames (6.2 frames NKI equivalent)
PS3 Turbo 3, PS360 PCB: 4.3-6.2 frames. Average 5.3 frames (5.8 frames NKI equivalent)
PS3 Turbo 4, PS360 PCB: 4.1-6.2 frames. Average 5.3 frames (5.8 frames NKI equivalent)

did u test GGPO?

Also FBA with sound disabled gives -2 frames of delay, so probably its very close to arcade

Emulators are on the todo list, along with CCC2 and Dreamcast.

Got some Dreamcast testing in.

All testing with MC Cthulu 2.5.4 1ms firmware.
Dreamcast Turbo 0: 4-5 frames of lag. 4.5 average - 5 f NKI equivalent
Dreamcast Turbo 1: 3-5 frames of lag, 4.2 average
Dreamcast Turbo 2: 3-5 frames of lag, 4.2 average
Dreamcast Turbo 3: 3-5 frames of lag, 4.1 average.
Dreamcast Turbo 4: 3-5 frames of lag, 4.1 average.
Dreamcast Turbo 5: 3-5 frames of lag, 3.9 average.
Dreamcast Turbo 6: 3-4 frames of lag, 3.5 average.

Also, Dreamcast turbo settings are faster than HDR turbo settings.

So basically the Dreamcast version, which was the tournament standard for ST for years, has the same amount of input lag as HDR on the 360, yet somehow HDR or HDR Classic mode (by far the easiest way to set up ST at tournaments adding no additional space/equipment requiments) isn’t considered a viable option by a select, vocal few. While at the same time 99% of people playing ST/HDR are doing it on either consoles or on laggy emulators (with wack speed settings). It’s very clear to me why our favorite game/s are quickly becoming more and more niche. It’s the only one where you can’t play and practice your game under the same conditions, with the same stick/controller, on the same equipment as you will see at a tourney.

The set-up can add significant lag. I expect that using the “Evo Monitor” and a MadCatz SE/TE stick instead of a CRT and a PS360 will add significant lag. FWIW, going from a good supergun to PS3 HDR is pretty jarring.

Dreamcast Frame Skip (short samples…)
Appears to be patterns of 32 in-game frames. The numbers are visible consecutive frames between skipped frames.

Turbo 6: 1,1,1,1,1,1,1…
Turbo 5: 2,2,1 (repeat)
Turbo 4: 3,3,2,3,2,3,2,3,2 (repeat)
Turbo 3: 3,3,3,3,3,3,.
Turbo 2: 4,4,3,4,3,4,3 (repeat)
Turbo 1:4,4,5 (repeat)
Turbo 0: No skip

Turbo settings correspond to 0,6,7,8,9,12, and 16 skipped frames per 32 visible frames.
Turbo 0 - 100% speed
Turbo 1 - 123% speed - approximately HDR turbo 2.
Turbo 2 - 128% speed - approximately HDR turbo 4.
Turbo 3 - 133% speed
Turbo 4 - 139% speed
Turbo 5 - 160% speed
Turbo 6 - 200% speed

Rufus: It’s great that you’re providing more quantitative results for input lag but is there any reason why the official DC controller or stick wasn’t tested instead? I’d feel a lot more confident with results using that that than with the unofficial MC Cthulhu setup.

Anyway, I’m not sure what “approximately” means but HDR speeds aren’t equivalent to DC speeds at any level except Turbo 0. My DC-HDR timeline in the first post has the details and were based on side-by-side comparison of the ending time of rounds. Based on that, DC T1 should be closer to HDR T3 speed. Feel free to use the same test and see for yourself.

HDR on Dreamcast, you mean the port of ST? I thought at the time that had been documented as a flawed conversion, although nowhere near as bad as the oddball Saturn rev found on the collection with Zero2’?

What about running Anniversary Edition on the old Xbox, choosing purely ST chars? HDR on 360 is something else entirely… just look at the Sagat juggles :wasted:

I would happily pay $25 right now for a pure and simple arcade port of Super Turbo to play (with acceptable network and lobby options) over Xbox Live.

In terms of game frames/shown frames
The speeds on the DC are 32/x (32/32, 32/26, 32/25, 32/24, 32/23, 32/20, and 32/16 for turbo 0-6)
On HDR it’s x/32 (32/32, 38/32, 39/32, 40/32, and 41/32 for turbo 0-4)
These are relative to turbo 0 based on the differemces in frame skip.

I used the mc cthulu because it was easier for wiring than the DC pad.

Here’s a better speed chart:

    Turbo Game Frames Shown Frames Speed
DC  0    32          32          1.00
    1    32          26          1.23
    2    32          25          1.28
    3    32          24          1.33
    4    32          23          1.39
    5    32          20          1.60
    6    32          16          2.00
HDR 0    32          32          1.00
    1    38          32          1.19
    2    39          32          1.22
    3    40          32          1.25
    4    41          32          1.28

Edit: Frameskip tested on Toodles disk ST.

Dangit. I just started playing HDR again but on the PS3 with a usb converter for my 360 Madcatz stick. lol Looks like I could not have picked a worse combination.

I know I’m late to the party, but I just wanted to chime in. Since there have been previous posts on this forum about which version is used in tourneys, I have a gripe about this whole 25th Anniversary tourney that’s going on. Why is it that Capcom tells everyone that HD REMIX is an official tourney game, yet they play it on classic mode. That doesn’t make sense. You might was well run it on ST cabinets or whatever has been done in the past. Instead the tourney is run under HD Remix, but the old character sprites are used and you have to play classic. Doesn’t that defeat the whole purpose of playing HD Remix in a tourney in the first place? I also wish they would have made this tourney more available to other areas in the country (US). Not everyone lives in major cities. Just seems like it would be a good idea to give everyone a fair shot vs having to drive x amount of miles to play.

I totally agree on this. I’m a big fan of both ST and HDR, but I must say HDR is way easier to setup and play. People who are advocates of GGPO, Supercade, etc… must realize that these programs are not the easiest to run. I’ve tried numerous times to run GGPO, but I takes too much time and effort for it to run properly. I always have lag issues, rooms wont load, I cant accept challenges, etc. So what do I do? Answer: say to heck with it and turn on HDR and get 20 games in. I just wish more ST players would return to HDR. There IS a classic mode on there for a reason. It may not be mirror perfect, but is it really enough to scare away players? Just sayin…

If u cant set up GGPO properly u dont deserve playing it

Freeway i told you, stop using DHCP, change your ip address to static, that’s your problem with GGPO, always was, always will be!

Dreamcast ST is so awesome that I find little reason to buy the actual cps2.
What capcom did with this gem, they did it awesome and with added dipswitches.

I just tried ShmupMAME and the difference coming from Final Burn Alpha used on GGPO is like night and day. There’s basically no input delay with ShmupMAME, compared to horrible input lag (which we’ve known about for years) with FBA. The thing with ShmupMAME is it’s been designed to remove as much input delay as possible, focusing especially on shmups, but they’ve also apparently made fixes to CPS2 drivers which allow games like ST to run very smoothly and feel extremely responsive.

I know moving to some version of MAME for GGPO is not happening ever, but it’s pretty disheartening when you switch between ShmupMAME and GGPO-FBA and feel the huge difference. I’m not sure if the latest versions of FBA have incorporated any input delay fixes for CPS2/ST, but I doubt it. We might be stuck using this outdated emulator forever.

Works perfectly for me using wired gigabit with a TPlink tl-wdr4300 router. DHCP is on. Setup for ports was super easy. My IP never changes on DHCP.

Usually people having connection problems are the ones with bad routers.